Deep Transverse Friction Massage For Treating Lateral Elbow or Lateral Knee Tendinitis

Cochrane Database of Systematic Reviews
Deep transverse friction massage for treating lateral elbow or

lateral knee tendinitis (Review)
Loew LM, Brosseau L, Tugwell P, Wells GA, Welch V, Shea B, Poitras S, De Angelis G, Rahman P
Loew LM, Brosseau L, Tugwell P, Wells GA, Welch V, Shea B, Poitras S, De Angelis G, Rahman P.
Deep transverse friction massage for treating lateral elbow or lateral knee tendinitis.
Cochrane Database of Systematic Reviews 2014, Issue 11. Art. No.: CD003528.
DOI: 10.1002/14651858.CD003528.pub2.
www.cochranelibrary.com
Deep transverse friction massage for treating lateral elbow or lateral knee tendinitis (Review)
Copyright 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SUMMARY OF FINDINGS FOR THE MAIN COMPARISON . . . . . . . . . . . . . . . . . . .
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 1.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 2.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 3.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ADDITIONAL SUMMARY OF FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AUTHORS CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ACKNOWLEDGEMENTS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.1. Comparison 1 Massage + therapeutic ultrasound (US) and placebo ointment vs US + placebo ointment only,
Outcome 1 Pain (VAS 0-100, 0 = worst) (change from baseline). . . . . . . . . . . . . . . . .
Outcome 2 Grip strength (ratio index, higher is better). . . . . . . . . . . . . . . . . . . .
Outcome 3 Function (VAS 0-100, 0 = worst). . . . . . . . . . . . . . . . . . . . . . . .
Outcome 4 Function (pain-free function; average number of pain-free items; higher is better). . . . . . .
Outcome 5 Functional status (number of successes to perform strengthening program). . . . . . . . .
Analysis 2.1. Comparison 2 Massage + phonophoresis vs phonophoresis only, Outcome 1 Pain (VAS 0-100, 0 = worst).
Analysis 2.2. Comparison 2 Massage + phonophoresis vs phonophoresis only, Outcome 2 Grip strength (ratio index, higher
is better). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.3. Comparison 2 Massage + phonophoresis vs phonophoresis only, Outcome 3 Function (VAS 0-100, 0 =
worst). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.4. Comparison 2 Massage + phonophoresis vs phonophoresis only, Outcome 4 Function (pain-free function;
average number of pain-free items; higher is better). . . . . . . . . . . . . . . . . . . . . .
Analysis 2.5. Comparison 2 Massage + phonophoresis vs phonophoresis only, Outcome 5 Functional status (number of
successes to perform strengthening program). . . . . . . . . . . . . . . . . . . . . . . .
Analysis 3.1. Comparison 3 Massage + physical therapy vs physical therapy only, Outcome 1 Pain. . . . . . . .
ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
WHATS NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . .
INDEX TERMS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
1
2
4
6
6
6
8
10
11
12
14
19
20
20
21
24
31
32
33
33
34
34
35
35
36
36
37
37
37
39
42
42
42
42
43
43
[Intervention Review]
Deep transverse friction massage for treating lateral elbow or

lateral knee tendinitis
Laurianne M Loew1 , Lucie Brosseau1 , Peter Tugwell2 , George A Wells3 , Vivian Welch4 , Beverley Shea3 , Stephane Poitras1 , Gino De
Angelis3 , Prinon Rahman1
1 School
of Rehabilitation Sciences, Faculty of Health Sciences, University of Ottawa, Ottawa, Canada. 2 Department of Medicine,
Faculty of Medicine, University of Ottawa, Ottawa, Canada. 3 Department of Epidemiology and Community Medicine, University of
Ottawa, Ottawa, Canada. 4 Bruyre Research Institute, University of Ottawa, Ottawa, Canada
Contact address: Lucie Brosseau, School of Rehabilitation Sciences, Faculty of Health Sciences, University of Ottawa, 451 Smyth Road,
Ottawa, ON, K1H 8M5, Canada. Lucie.Brosseau@uottawa.ca.
Editorial group: Cochrane Musculoskeletal Group.
Publication status and date: New search for studies and content updated (no change to conclusions), published in Issue 11, 2014.
Review content assessed as up-to-date: 1 July 2014.
Citation: Loew LM, Brosseau L, Tugwell P, Wells GA, Welch V, Shea B, Poitras S, De Angelis G, Rahman P. Deep transverse friction
massage for treating lateral elbow or lateral knee tendinitis. Cochrane Database of Systematic Reviews 2014, Issue 11. Art. No.: CD003528.
DOI: 10.1002/14651858.CD003528.pub2.
ABSTRACT
Background
Deep transverse friction massage, one of several physical therapy interventions suggested for the management of tendinitis pain, was
first demonstrated in the 1930s by Dr James Cyriax, a renowned orthopedic surgeon in England. Its goal is to prevent abnormal fibrous
adhesions and abnormal scarring. This is an update of a Cochrane review first published in 2001.
Objectives
To assess the benefits and harms of deep transverse friction massage for treating lateral elbow or lateral knee tendinitis.
Search methods
We searched the following electronic databases: the specialized central registry of the Cochrane Field of Physical and Related Therapies,
the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, the Cumulative Index to Nursing and Allied
Health Literature (CINAHL), Clinicaltrials.gov, and the Physiotherapy Evidence Database (PEDro), up until July 2014. The reference
lists of these trials were consulted for additional studies.
Selection criteria
All randomized controlled trials (RCTs) and controlled clinical trials (CCTs) comparing deep transverse friction massage with control
or other active interventions for study participants with two eligible types of tendinitis (ie, extensor carpi radialis tendinitis (lateral
elbow tendinitis, tennis elbow or lateral epicondylitis or lateralis epicondylitis humeri) and iliotibial band friction syndrome (lateral
knee tendinitis)) were selected. Only studies published in English and French languages were included.
Data collection and analysis
Two review authors independently assessed the studies on the basis of inclusion and exclusion criteria. Results of individual trials were
extracted from the included study using extraction forms prepared by two independent review authors before the review was begun.
Data were cross-checked by a third review author. Risk of bias of the included studies was assessed using the Risk of bias tool of The
Cochrane Collaboration. A pooled analysis was performed using mean difference (MD) for continuous outcomes and risk ratio (RR)
for dichotomous outcomes with 95% confidence intervals (CIs).
Main results
Two RCTs (no new additional studies in this update) with 57 participants met the inclusion criteria. These studies demonstrated high
risk of performance and detection bias, and the risk of selection, attrition, and reporting bias was unclear.
The first study included 40 participants with lateral elbow tendinitis and compared (1) deep transverse friction massage combined
with therapeutic ultrasound and placebo ointment (n = 11) versus therapeutic ultrasound and placebo ointment only (n = 9) and (2)
deep transverse friction massage combined with phonophoresis (n = 10) versus phonophoresis only (n = 10). No statistically significant
differences were reported within five weeks for mean change in pain on a 0 to 100 visual analog scale (VAS) (MD -6.60, 95% CI -28.60
to 15.40; 7% absolute improvement), grip strength measured in kilograms of force (MD 0.10, 95% CI -0.16 to 0.36) and function on
a 0 to 100 VAS (MD -1.80, 95% CI -0.18.64 to 15.04; 2% improvement), pain-free function index measured as the number of painfree items (MD 1.10, 95% CI -1.00 to 3.20) and functional status (RR 3.3, 95% CI 0.4 to 24.3) for deep transverse friction massage,
and therapeutic ultrasound and placebo ointment compared with therapeutic ultrasound and placebo ointment only. Likewise for deep
transverse friction massage and phonophoresis compared with phonophoresis alone, no statistically significant differences were found
for pain (MD -1.2, 95% CI -20.24 to 17.84; 1% improvement), grip strength (MD -0.20, 95% CI -0.46 to 0.06) and function (MD
3.70, 95% CI -14.13 to 21.53; 4% improvement). In addition, the GRADE (Grades of Recommendation, Assessment, Development
and Evaluation) approach was used to evaluate the quality of evidence for the pain outcome, which received a score of very low.
Pain relief of 30% or greater, quality of life, patient global assessment, adverse events, and withdrawals due to adverse events were not
assessed or reported.
The second study included 17 participants with iliotibial band friction syndrome (knee tendinitis) and compared deep transverse
friction massage with physical therapy intervention versus physical therapy intervention alone, at two weeks. Deep transverse friction
massage with physical therapy intervention showed no statistically significant differences in the three measures of pain relief on a 0 to
10 VAS when compared with physical therapy alone: daily pain (MD -0.40, 95% CI -0.80 to -0.00; absolute improvement 4%), pain
while running (scale from 0 to 150) (MD -3.00, 95% CI -11.08 to 5.08), and percentage of maximum pain while running (MD 0.10, 95% CI -3.97 to 3.77). For the pain outcome, absolute improvement showed a 4% reduction in pain. However, the quality of
the body of evidence received a grade of very low. Pain relief of 30% or greater, function, quality of life, patient global assessment of
success, adverse events, and withdrawals due to adverse events were not assessed or reported.
Authors conclusions
We do not have sufficient evidence to determine the effects of deep transverse friction on pain, improvement in grip strength, and
functional status for patients with lateral elbow tendinitis or knee tendinitis, as no evidence of clinically important benefits was found.
The confidence intervals of the estimate of effects overlapped the null value for deep transverse friction massage in combination with
physical therapy compared with physical therapy alone in the treatment of lateral elbow tendinitis and knee tendinitis. These conclusions
are limited by the small sample size of the included randomized controlled trials. Future trials, utilizing specific methods and adequate
sample sizes, are needed before conclusions can be drawn regarding the specific effects of deep transverse friction massage on lateral
elbow tendinitis.
PLAIN LANGUAGE SUMMARY

Deep transverse friction massage for the treatment of lateral elbow or lateral knee tendinitis
Review question
We conducted an update of the review of the effects of deep transverse friction massage (DTFM) for people with lateral elbow or knee
tendinitis. We found two studies (no new additional studies in this update) with 57 people.
Background: what is tendinitis and what is deep tendon friction massage?
Tendons are fibrous structures that attach muscle to bone. Tendinitis essentially is seen when a tendon suffers from inflammation
(painful swelling). The inflammation is due to too much stress on the tendon caused by performing repetitive movements. This causes
pain and stiffness in the joint such as the elbow or the knee.
Deep transverse friction massage (DTFM) is a physical therapy technique often used to reduce damage and scarring caused by
inflammation. It increases blood flow to the joint, which facilitates healing of the tendon by increasing the supply of oxygen transported
to the injury.
Study characteristics
One study with a duration of five weeks reviewed (1) the effects of deep transverse friction massage combined with therapeutic ultrasound
and placebo ointment compared with therapeutic ultrasound and placebo ointment in 20 people with lateral elbow tendinitis (tennis
elbow), as well as (2) the effects of deep transverse friction massage combined with phonophoresis compared with phonophoresis alone
in 20 people with lateral elbow tendinitis (tennis elbow). The other study, with a duration of two weeks, reviewed the effects of deep
transverse friction massage with physical therapy intervention compared with physical therapy intervention alone in 17 people with
lateral knee tendinitis.
Key results
What happens to people with lateral elbow tendinitis (tennis elbow) who are treated with deep transverse friction massage?
We are uncertain whether deep transverse friction massage improves pain and function (very low-quality evidence).
No studies reported pain relief of 30% or greater, quality of life, patient global assessment, adverse events, and withdrawals due to
adverse events.
We often do not have precise information about side effects and complications. This is particularly true for rare but serious side effects.
What happens to people with lateral knee tendinitis who are treated with deep transverse friction massage?
We are uncertain whether deep transverse friction massage improves pain (very low-quality evidence).
No studies reported pain relief of 30% or greater, function, quality of life, patient global assessment, adverse events, and withdrawals
due to adverse events.
We often do not have precise information about side effects and complications. This is particularly true for rare but serious side effects.
S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
Massage + therapeutic ultrasound and placebo ointment compared with ultrasound + placebo ointment only (follow-up 2 weeks) for treating tendinitis
Patient or population: patients with extensor carpi radialis tendinitis
Settings: community sports injuries clinic in Canada
Intervention: massage + therapeutic ultrasound and placebo ointment
Comparison: therapeutic ultrasound + placebo ointment only
Outcomes
Illustrative comparative risks* (95% CI)
Assumed risk
Relative effect
(95% CI)
Number of participants
(studies)
Quality of the evidence

(GRADE)
Comments
20
(1 study)

Very lowa,b,c
MD = -6.60 (-28.60 to
15.40)
Absolute improvement =
-7% (-29% to -15%)
Relative
percentage change = 8%
(-24% to 37%)
Not statistically significant
See comment
Not measured
20
(1 study)

Very lowa,b,c
MD = -1.80 (-18.64 to
15.04)
2% (-19% to 15%)
Relative
percentage
change = -3% (-28% to
Corresponding risk
Control (US + placebo Massage + US and

ointment only)
placebo ointment
Pain
Mean change in pain in
Visual analog scale
the control groups was
Scale from 0 to 100 4.6 mm
(lower is better)
Follow-up: mean 2 weeks

the intervention groups
was
6.6 lower
(15.40 lower to 28.60
higher)
Proportion
report- See comment
ing pain relief of 30% or
greater not measured
See comment
Function
Visual analog function index
Scale from 0 to 100
(higher is better)
Mean function (VAS 0100, 0 = worst) in the

intervention groups was
1.8 lower
(18.64 lower to 15.04
higher)
Mean function (vas 0 to

100, 0 = worst) in the
control groups was
78.1 mm
Not estimable
23%)
Quality of life-not mea- See comment
sured
See comment
Not estimable
See comment
Not measured
Patient global assess- See comment

ment of success not
measured
See comment
Not estimable
See comment
Not measured
Adverse events not mea- See comment

sured
See comment
Not estimable
See comment
Not measured
Withdrawals due to ad- See comment

verse events not measured
See comment
Not estimable
See comment
Not measured
*The basis for the assumed risk (e.g. median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed
risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; ECRT: Extensor carpi radialis tendinitis; RR: Risk ratio.
GRADE Working Group grades of evidence.
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.
a
Allocation concealment was unclear. Only assessors were blinded. Baseline imbalance was reported.
few participants.
c Wide confidence intervals.
b Very
BACKGROUND
Description of the condition

Extensor carpi radialis tendinitis (ECRT) (ie, lateral elbow tendinitis, tennis elbow or lateral epicondylitis, or lateralis epicondylitis humeri) is a local inflammation near the proximal (upper) attachments of wrist extensor muscles, characterized by pain on palpation of the lateral epicondyle of the humerus and in resisted
movements against wrist extension (Struijs 2002). The prevalence
of lateral elbow tendinitis (tennis elbow) varies between 1% and
10% of the population; it commonly occurs between the ages of
34 and 74 years (Allander 1974; Roquelaure 2006; Shiri 2006;
Walker-Bone 2004). Stratford 1989 reported that lateral elbow
tendinitis (tennis elbow) does not seem to be a degenerative condition, as its prevalence declines after the age of 42 (Allander 1974).
It is a syndrome of overuse (eg, use of computer mouse, racquet
sports) that can result in considerable socioeconomic costs resulting from prolonged leave of absence from work (Struijs 2002).
Iliotibial band friction syndrome (ITBFS) (lateral knee tendinitis) is an overuse musculoskeletal injury that causes lateral knee
pain and is frequently observed in long distance runners, cyclists,
football players, and military personnel. The incidence of lateral
knee tendinitis varies between 1.6% and 52%, depending upon
the population studied (Jordaan 1994; Kirk 2000; Lavine 2010;
Pinshaw 1984; Strauss 2011). The mechanism of lateral knee tendinitis appears to be repetitive friction of the iliotibial band moving over the lateral femoral condyle during knee flexion/extension
(Schwellnus 1991). The cause of ITBFS (lateral knee tendinitis)
is multifactorial (Messier 1988). Three main risk factors play a
large role in the development of lateral knee tendinitis, according
to the current literature: (1) biomechanical factors such as muscle imbalance causing maximal braking forces required to control
body movements depending on velocity and angle of inclination
(Messier 1995), (2) anthropometric factors such as abnormalities
in leg and/or feet anatomy (Messier 1988), and (3) training factors such as inadequate warm-up and cool-down training (Messier
1988; Messier 1995).
2002; Thaunton 1987; Wilk 2004); (5) prevention of reinjury

(Hart 1994; Reinold 2002); and (6) gradual return to training
(Reinold 2002; Thaunton 1987). The surgical approach includes
resection of the impinging portion of the extensor carpi radialis
brevis tendon release (Grewal 2009) or the iliotibial band (Kirk
2000; Martens 1989).
Deep transverse friction massage (DTFM) is a technique popularized by Dr James Cyriax (Cyriax 1975a; Cyriax 1975b) for relief of
pain and inflammation in musculoskeletal conditions. Deep friction is given transverse to the fiber direction of involved tissue, and
friction massage must be sufficiently deep to impact the affected
structure.
How the intervention might work

Deep transverse friction massage is offered in the treatment of various musculoskeletal conditions. This technique attempts to prevent or destroy abnormal fibrous adhesions (cross-links or crossbridges) by imposing stress transversely to the remodeling collagen
of the tissue to soften the adhesion. Thus, DTFM also optimizes
the quality of the scar tissue by realigning the collagen of normal
soft tissue fibers in a longitudinal way (Walker 1984). It has been
indicated that DTFM enhances normal healing conditions and
prevents abnormal scarring. Its mechanical action causes hyperemia (ie, an increase in blood flow to the affected area).
Why it is important to do this review

No meta-analyses or literature reviews have reported the efficacy of
this type of massage for lateral elbow or knee tendinitis. Clinicians
and patients must be provided with evidence that will enable them
to make informed decisions regarding treatment options.
OBJECTIVES
To assess the benefits and harms of deep transverse friction massage
for treating lateral elbow or lateral knee tendinitis.
Description of the intervention

Treatment of tendinitis consists of medical, physiotherapeutic,
and surgical approaches. The medical approach encompasses rest
and the prescription of anti-inflammatory agents combined with
anti-inflammatory or analgesic medication to control pain and
inflammation (Kirk 2000). Goals of the physiotherapeutic approach in the treatment of tendinitis include (1) control of pain
and inflammation (Cyriax 1975a; Cyriax 1975b; Thaunton 1987;
Wilk 2004); (2) correction of biomechanical deficiencies (Johnson
2007; Thaunton 1987); (3) restoration of motion (Hart 1994); (4)
increase in strength, endurance, and function (Hart 1994; Reinold
METHODS
Criteria for considering studies for this review

Types of studies
According to an a priori protocol, randomized controlled trials
(RCTs) and controlled clinical trials (CCTs) (and CCTs with pseudorandom methods of allocating participants to treatment) were
selected because they are defined as comparative controlled trials

with the presence of comparison groups, which allow evaluation
of DTFM and patients with tendinitis pain compared with other
types of interventions and populations. Only articles published in
the English language or the French language were included.
Types of participants
Studies were included if study participants were 18 years of age
or older and had been diagnosed with tendinitis (lateral elbow
tendinitis and knee tendinitis). Participants had to show no signs
of psychiatric conditions and had to demonstrate stable physical
and medical status. Mixed populations (individuals with multiple
conditions other than tendinitis) were not permitted. Inclusion
criteria for lateral elbow tendinitis comprised (1) tenderness on
palpation over the lateral aspect of the elbow and (2) pain in the
lateral aspect of the elbow during resisted wrist extension. For
knee tendinitis, inclusion criteria consisted of (1) history of pain
on the lateral aspect of the knee during running, (2) tenderness
over the lateral femoral condyle at rest, and (3) aggravation of
symptoms at 30 degrees of knee flexion. For the complete list of
exclusion criteria and the PICOTS (populations, interventions,
comparators, outcomes, timing, and setting) framework, see Table
1.
Types of interventions
Trials comparing deep transverse friction massage versus no therapy or active treatments (such as therapeutic ultrasound, exercise
program, or phonophoresis therapy) were accepted, provided they
were given equally to all treatment groups. So trials comparing
DTFM combined with other physical therapy manipulative interventions were excluded if co-interventions were not applied to
the control group (ie, if investigators wished to assess the effects
of deep massage therapy alone).
Types of outcome measures
Major outcomes
Mean (or mean change in) pain.

Proportion reporting pain relief of 30% or greater.
Function.
Quality of life.
Patient global assessment of success.
Total number of withdrawals due to adverse effects.
Number of adverse events.
Minor outcomes
In addition to these outcomes, one of the review authors (LB) has

developed a theoretical framework for important outcome measures for physical therapy interventions (Morin 1996). These outcomes were considered as secondary measures of effectiveness and
include the following.
Joint range of motion (ROM).
Muscle strength (and grip strength for tendenitis of upper
extremities).
Endurance.
Outcomes were assessed at end of treatment or at end of followup.
Search methods for identification of studies
Electronic searches
We searched the specialized register of the Cochrane Field of Physical and Related Therapies, the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, the Cumulative Index to Nursing and Allied Health Literature (CINAHL),
Clinicaltrials.gov, and the Physiotherapy Evidence Database (PEDro), until July 2014 (Figure 1). The search included deep transverse friction therapy terms and tendinitis pain terms. Full strategies for these databases are provided in Appendix 1 (MEDLINE),
Appendix 2 (EMBASE), Appendix 3 (CENTRAL), Appendix 4
(CINAHL), Appendix 5 (PEDro), and Appendix 6 (Clinicaltrials.gov).
Figure 1. Study flow diagram.
Searching other resources

We searched the reference lists of selected articles.
Data collection and analysis
Measures of treatment effect

Mean differences (MDs) between experimental and control groups
were calculated for continuous data, with the same measurements
and standardized mean differences (SMDs) for continuous outcomes with different measurements, and 95% confidence intervals
(CIs). For dichotomous data, risk ratios (RRs) and 95% CIs were
calculated.
Selection of studies
Two independent review authors (LB, LML) examined the titles
and abstracts of the trials identified by the search strategy to select
trials that met the inclusion criteria (see Table 1 for more details).
All trials classified as relevant by at least one of the review authors
were retrieved as full-text articles and were assessed for eligibility.
Disagreements were resolved by consensus. When selecting studies
for inclusion, we did not consider languages other than French
and English.
Individual participants represented the unit of analysis in these

studies. No cluster-randomized controlled trials or cross-over design trials were selected. In cases of multiple treatment groups, we
made multiple pairwise comparisons between all possible pairs of
intervention groups to avoid double-counting of participants in
the meta-analyses.
Data extraction and management
Dealing with missing data
The results of the individual trials were extracted from the included study by two independent review authors (LB, LML) using predetermined extraction forms. Data were cross-checked by
a third review author (BS). The extraction forms were developed
and pilot-tested on the basis of other forms used by the Cochrane
Musculoskeletal Review Group. The extraction form documented
specific information about DTFM, including (1) characteristics
of the technique and (2) methods of therapeutic application such
as duration, frequency, rhythm, pressure, and total number of sessions. Also, forms were designed to collect information about participant characteristics (eg, age, gender, injury type, injury duration), comparator intervention characteristics, and outcomes (including scale of tool, direction of effect [eg, lower score = worse
off ]). Final data values were based on consensus of the two review
authors. In the event of multiple time points, we would collect the
last time point.
The procedure was to contact the study authors if only an abstract

was provided, or if available results were not sufficient for the data
analysis, or to identify additional studies. For this review update,
it was not necessary to contact the study authors.
Assessment of risk of bias in included studies

Two independent review authors (LB, LML) assessed risk of bias
using the Risk of bias tool of The Cochrane Collaboration
(Higgins 2011); differences were resolved by consensus. This tool
assesses seven domains: sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome
assessment, incomplete outcome data, selective outcome reporting, and other issues. For other issues, we considered baseline
imbalance as a potential source of bias. The review authors classified the different items for the two included studies as high risk,
unclear risk, or low risk of bias. A third review author (BS)
was consulted when necessary.
Unit of analysis issues
Assessment of heterogeneity
We planned to assess heterogeneity using the I2 statistic when a
value of 0% to 40% might not be important, 30% to 60% may
represent moderate heterogeneity, 50% to 90% may represent
substantial heterogeneity, and 75% to 100% represents considerable heterogeneity (Higgins 2011).
Assessment of reporting biases

We intended to assess the possibility of publication bias by using
funnel plots.
We also planned to assess trial protocols versus published reports.
For example, for studies published after July 1, 2005, we would
screen the Clinical Trial Register at the International Clinical Trials Registry Platform of the World Health Organization (http://
apps.who.int/trialssearch) for the a priori trial protocol. We would
evaluate whether selective reporting of outcomes is present (outcome reporting bias).
We planned to compare the fixed-effect estimate against the random-effects model to assess the possible presence of small sample bias (ie, in which the intervention effect is more beneficial
in smaller studies) in the published literature. In the presence of
small sample bias, the random-effects estimate of the intervention
is more beneficial than the fixed-effect estimate (Higgins 2011).
Data synthesis
We planned to pool the intervention versus a common comparator by using the fixed-effect model of meta-analysis if data were
homogenous, and the random-effects model if heterogeneity was
substantive (ie, I2 > 50%).
Subgroup analysis and investigation of heterogeneity

No studies were pooled; therefore exploration of heterogeneity and
subgroup analyses were not possible.
Sensitivity analysis
We planned sensitivity analyses based on the risk of bias assessment of included studies (allocation concealment, blinding of participants, intention-to-treat analysis). However, we could not perform sensitivity analyses for this review.
Grading of the evidence and summary of findings

tables
We graded the overall quality of the evidence for each outcome
using the GRADE (Grading of Recommendations Assessment,
Development and Evaluation) approach (Guyatt 2008), as recommended by The Cochrane Collaboration (Schnemann 2011a).
This approach consists of four levels of quality: high, moderate,
low, and very low. Evidence-based randomized studies start at a
high quality rating and could be downgraded on the basis of five
factors (study limitations, inconsistency of results, indirectness of
evidence, imprecision, and publication bias).
We incorporated grading of evidence into the Summary of findings table. We also presented other key results in the Summary of
findings table, as recommended by The Cochrane Collaboration
(Schnemann 2011b), including the absolute and relative magnitude of effect of the interventions examined. Main outcomes in
the Summary of findings table were mean (or mean change in)
pain, proportion reporting pain relief of 30% or greater, function,
quality of life, patient global assessment of success, total number of
withdrawals due to adverse effects, and number of adverse events.
RESULTS
Description of studies
Results of the search

We retrieved 853 references after deduplication and identified 34
potentially eligible articles through electronic database searches
(Figure 1). Two review authors evaluated the studies and screened
28 full-text articles on deep transverse friction massage and pain
of tendinitis for eligibility. Two full-text articles met our inclusion
criteria, and 26 were excluded.
Included studies
Two trials met the inclusion criteria. See the Characteristics of
included studies table.
In the first included RCT (Stratford 1989), all participants were
18 years of age or older. Mean participant age was 43.3 years, and
symptom duration was 4.25 weeks. This included RCT examined
the efficacy of DTFM in the management of extensor carpi radialis tendinitis (lateral elbow tendinitis or tennis elbow) and included the following comparison groups: (1) deep transverse friction massage combined with therapeutic ultrasound and placebo
ointment (n = 11) versus therapeutic ultrasound combined with
placebo ointment (n = 9) and (2) deep transverse friction massage
and phonophoresis (n = 10) versus phonophoresis alone (n = 10).
The study consisted of nine treatment sessions within five weeks.
The second RCT (Schwellnus 1992) compared two groups: One
received combined physical therapy interventions with DTFM,
and the other received combined physical therapy interventions
without DTFM. The study consisted of four consecutive treatment sessions in two weeks. A total of 17 participants with iliotibial band friction syndrome (lateral knee tendinitis) were randomly
assigned. All participants in this study were prescribed rest, stretching exercises, cryotherapy and therapeutic ultrasound. Mean participant age was 27.6 years, disease duration was 48.5 weeks, years
of running experience were 6.6, and the weekly distance in kilometers was 54.5 for both groups. Injury severity was assessed by
grading the pain level (ie, grade 1: pain experienced after running;
grade 2: pain experienced during running; grade 3: pain experienced during running associated with restriction of distance or
velocity; grade 4: severe pain that prevents running). The grade of
injury was 3.4 of a maximum possible of 4 for both groups.
Excluded studies
Twenty-six trials were excluded for the reasons outlined here: (1)
healthy participants (Chiarello 1997; Crosman 1984), (2) not tendinitis (Balke 1989; Feehan 1989; Thomee 1997; Zhang 1987),
(3) combined modalities (Pellecchia 1994), (4) combination of interventions and co-interventions not applied to the control group
(Baltaci 2001; Bisset 2007; Fernandez 2006; Fernandez 2008;
Nagrale 2009; Smidt 2002; Stasinopoulos 2004a; Stasinopoulos
2006; Struijs 2006; Viswas 2012), (5) no control goup (Kohia
2008; Malier 1986; Mayer 2007; Stasinopoulos 2004b; Verhaar
1995), (6) pilot design (Struijs 2003), and (8) not an RCT
(Iwatsuki 2001; Joseph 2012; Zheng 2012). See the Characteristics
of excluded studies table.
10
Risk of bias in included studies

We assessed the risk of bias for each included study using the
Risk of bias tool of The Cochrane Collaboration. The risk of
bias graph and summary are presented in Figure 2 and Figure 3,
respectively.
Figure 2. Risk of bias graph: review authors judgements about each risk of bias item presented as
percentages across all included studies.
11
Figure 3. Risk of bias summary: review authors judgements about each risk of bias item for each included
study.
Allocation
In the RCT for extensor carpi radialis tendinitis (lateral elbow

tendinitis) (Stratford 1989), the study authors used a 2 2 factorial
design and created strata on the basis of pain-free grip strength.
Afterward, participants were randomly assigned to one of the four
treatment groups using a balanced blocked randomization table.
In the second RCT for iliotibial band friction syndrome (lateral
knee tendinitis) (Schwellnus 1992), participants were randomly
assigned, but the methods of sequence generation and allocation
were not described, and so risk was judged as unclear.
Blinding
Participants and personnel were not blinded to treatment. As this
is a physical intervention, it is more difficult to blind participants
and therapists. Only the outcome assessors were blinded to the
treatment groups of participants in both included studies. We did
not have enough information to assess the impact on patientreported outcomes, but detection bias is more likely high risk
for self-reported subjective outcomes (eg, pain).
Incomplete outcome data
12
One trial (Schwellnus 1992) reported a 15% dropout rate. Duration of follow-up was mentioned, but investigators did not perform an intention-to-treat analysis. In the second trial (Stratford
1989), withdrawals and dropouts were not mentioned. However,
all randomly assigned participants were accounted for in the results.
Selective reporting
Both studies reported planned outcomes whether or not findings
were significant (Schwellnus 1992; Stratford 1989). No protocols
were found.
Other potential sources of bias
Baseline imbalance was noted in both included studies (Schwellnus
1992; Stratford 1989). The duration of symptoms, for example,
was different in the intervention group compared with the control
group.
Effects of interventions
See: Summary of findings for the main comparison Massage
+ ultrasound and placebo ointment compared with ultrasound
+ placebo ointment only for treating lateral elbow tendinitis
(tennis elbow); Summary of findings 2 Massage + phonophoresis
compared with phonophoresis alone for treating lateral elbow
tendinitis (tennis elbow); Summary of findings 3 Deep transverse
massage + physical therapy compared with physical therapy alone
for treating lateral knee tendinitis
No studies were pooled; therefore assessment of heterogeneity and
subgroup analyses were not possible. (See Data and analyses section
for additional details.)
Deep transverse friction masssage + therapeutic
ultrasound and placebo ointment versus therapeutic
ultrasound and placebo ointment
The RCT for extensor carpi radialis tendinitis (tennis elbow)
(Stratford 1989) showed no statistically significant differences in
pain intensity, grip strength, and functional status after nine consecutive sessions of deep transverse friction massage combined
with therapeutic ultrasound and placebo ointment (n = 11) compared with therapeutic ultrasound alone (n = 9). We found the
following: mean change in pain on a 0 to 100 visual analog scale
(VAS) (MD -6.60, 95% CI -28.60 to 15.40) (Analysis 1.1), grip
strength in kilograms of force (MD 0.10, 95% CI -0.16 to 0.36)
(Analysis 1.2), function (VAS 0 to 100) (MD -1.80, 95% CI 18.64 to 15.04) (Analysis 1.3), pain-free function index measured
as the number of pain-free items (MD 1.10, 95% CI -1.00 to
3.20) (Analysis 1.4), and functional status assessed by the physician as the number of successes needed to perform the strengthening program (RR 3.3, 95% CI 0.4 to 24.3) (Analysis 1.5). This
difference was not clinically or statistically significant. The quality

of the body of evidence was very low according to the GRADE
approach.
Pain-free function scores recorded in the two groups at the end
of the study were -1.8 2.7 for deep transverse friction massage
combined with therapeutic ultrasound and placebo ointment and
2.7 2.1 for therapeutic ultrasound and placebo ointment. We
found the following values for the other measures, at the end of
the study.
Pain VAS: 44.8 33.4 for deep transverse friction massage
combined with therapeutic ultrasound and placebo ointment
and 23.8 17.0 for therapeutic ultrasound alone.
Function VAS: 76.3 21.9 for deep transverse friction
massage combined with therapeutic ultrasound and placebo
ointment and 78.1 16.5 for therapeutic ultrasound alone.
Absolute improvement measures demonstrated reduction in
pain by 7% and improvement in function by 2%.
As a minor outcome, the ratio index of pain-free grip
strength scores was as follows: 0.7 0.3 for deep transverse
friction massage combined with therapeutic ultrasound and
placebo ointment and 0.6 0.3 for therapeutic ultrasound alone.
Quality of life, patient global assessment of success, adverse
events, withdrawals due to adverse events, joint range of motion,
and endurance were not assessed or reported. (See Summary of
findings for the main comparison.)
Deep transverse friction masssage + phonophoresis

versus phonophoresis alone
Mean differences and risk ratios were also calculated for the comparison of deep transverse friction massage and phonophoresis (n
= 10) versus phonophoresis alone (n = 10) for pain on a 0 to 100
VAS scale (MD -1.2, 95% CI -20.24 to 17.84) (Analysis 2.1), grip
strength in kilograms of force (MD -0.20; 95% CI -0.46 to 0.06)
(Analysis 2.2), function (VAS 0 to 100) (MD 3.70, 95% CI 14.13 to 21.53) (Analysis 2.3), pain-free function index measured
as the number of pain-free items (MD 0.10, 95% CI -2.27 to
2.47) (Analysis 2.4), and functional status assessed by the physician as the number of successes needed to perform the strengthening program (RR 0.67, 95% CI 0.1 to 3.2) (Analysis 2.5) in
extensor carpi radialis tendinitis (Stratford 1989). This difference
was not clinically or statistically significant. Again, the quality of
the body of evidence received a score of very low according to
the GRADE approach.
The study authors reported different results for the two groups
for pain-free function scores, at the end of the study: 3.7 2.8
for deep transverse friction massage and phonophoresis compared
with 3.6 2.6 for phonophoresis alone.
Pain VAS: 24.6 20.6 for deep transverse friction massage
and phonophoresis compared with 21.8 30.4 for
phonophoresis alone.
Function VAS: 82.5 16.3 for deep transverse friction
13
massage and phonophoresis compared with 78.8 23.7 for

Absolute improvement confirmed reduction in pain by 1%
and improvement in function by 4%.
As a minor outcome, results for the ratio index of pain-free
grip strength were as follows: 0.5 0.3 for deep transverse
friction massage and phonophoresis compared with 0.7 0.3 for
Proportion reporting pain relief of 30% or greater, quality
of life, patient global assessment of success, adverse events,
withdrawals due to adverse events, joint range of motion, and
endurance were not assessed or reported. (See Summary of
findings 2.)
Deep transverse friction masssage and physical

therapy versus physical therapy alone
Only one small study with 17 participants looked at this comparison in iliotibial band friction syndrome (knee tendinitis)
(Schwellnus 1992). Although within-group reduction in the three
types of pain relief measured was significant (overall daily pain,
pain while running, and percentage of maximum pain when running) on a 0 to 10 VAS scale, in both treatment groups after
two weeks of follow-up (P value < 0.0005), no statistically significant between-group differences (mean differences (MDs)) were
reported for daily pain (MD -0.40, 95% CI -0.80 to 0.00), pain
while running measured as the total daily pain of all participants

while running (scale from 0 to 150) (MD -3.00, 95% CI -11.08
to 5.08), and percentage of maximum pain while running (MD
-0.10, 95% CI -3.97 to 3.77) (Analysis 3.1). The absolute improvement highlighted a reduction in pain by 4%. It is important
to note that the difference in daily pain was significant at P value
0.05 in favor of DTFM, although care in assessing this result must
be exercised, as (1) the result is borderline significant and (2) the
data are determined from a graph and are subject to interpretation
errors. In addition, the quality of the body of evidence according
to the GRADE approach received a score of very low.
Mean daily pain scores recorded in the two groups, at the
end of the study, were as follows: 0.6 0.3 for the experimental
group (combined physiotherapy interventions and DTFM) and
1 0.5 for the control group (physiotherapy interventions only).
At the same period, the study authors found different
results in the two groups for total pain while running: 20 9 for
the experimental group and 23 8 for the control group.
Finally, the percentage of maximum pain experienced
during running for the two groups was similar at the end of the
study: 8 5 for the experimental group and 8.1 3 for the
control group.
Proportion reporting pain relief of 30% or greater, function,
quality of life, patient global assessment of success, number of
adverse effects, total number of withdrawals due to adverse
effects, joint range of motion, muscle strength, and endurance
were not mentioned or reported. (See Summary of findings 3.)
14
A D D I T I O N A L S U M M A R Y O F F I N D I N G S [Explanation]
Massage + phonophoresis compared with control (phonophoresis only) (follow-up 2 weeks) for treating tendinitis
Patient or population: patients with extensor carpi radialis tendinitis
Settings: community sports injuries clinic in Canada
Intervention: massage + phonophoresis
Comparison: phonophoresis only
Outcomes
Assumed risk
Relative effect
(95% CI)
(studies)

(GRADE)
Comments
20
(1 study)

Very lowa,b,c
MD = -1.2 (-20.24 to 17.

84)
-1%
(-20% to 18%)
Relative
(-86% to 97%)
See comment
Not measured
20
(1 study)

Very lowa,b,c
MD = 3.70 (-14.13 to 21.

53)
4% (-14% to 22%)
Corresponding risk
Control (phonophore- Massage

sis only) (follow-up 2 phonophoresis
weeks)
15
Pain
Visual analog scale
the control groups was
Scale from 0 to 100 1 mm
(lower is better)

the intervention groups
was
1.2 lower
(17.84 lower to 20.24
higher)
Proportion
report- See comment
See comment
Function
Mean function in the conVisual analog scale
trol groups was
Scale from 0 to 100 78.8 mm
(higher is better)
Mean function in the intervention groups was

3.7 higher
(14.13 lower to 21.53
higher)
Not estimable
Relative
(-18% to 28%)
Quality of life not mea- See comment
sured
See comment
Not estimable
See comment
Not measured

ment of success not
measured
See comment
Not estimable
See comment
Not measured

sured
See comment
Not estimable
See comment
Not measured

verse events not measured
See comment
Not estimable
See comment
Not measured
CI: Confidence interval; ECRT: extensor carpi radialis tendinitis; RR: Risk ratio.
a Allocation
concealment was unclear. Only assessors were blinded. Baseline imbalance was reported.
few participants.
c Wide confidence interval.
b Very
16
Deep transverse friction massage + physical therapy compared with physical therapy alone
Patient or population: patients with iliotibial band friction syndrome (knee tendinitis)
Settings: community sports injury clinic in South Africa
Intervention: deep transverse massage and physical therapy
Comparison: physical therapy alone
Outcomes
Daily pain
Visual
analog scale (VAS)
Scale from 0 to
10 (lower is better)
Follow-up mean 2
weeks
Assumed risk
Corresponding risk
Control
Deep transverse massage
Relative effect
(95% CI)
Mean daily pain

Mean daily pain in
in the control groups was the intervention groups
1 point
was
0.4 lower
(0.8 lower to 0.00 higher)
(studies)

(GRADE)
Comments
17
(1 study)

very lowa,b
MD = -0.4 (-0.8 to 0)
-4% (-8% to 0%)
Relative
percentage
change = -40% (-80% to
0%)
Proportion
report- See comment
See comment
Not estimable
See comment
Not measured
Function not measured
See comment
See comment
Not estimable
See comment
Not measured
Quality of life not mea- See comment

sured
See comment
Not estimable
See comment
Not measured

ment not measured
See comment
Not estimable
See comment
Not measured
17

sured
See comment
Not estimable
See comment
Not measured

verse events
See comment
Not estimable
See comment
Not measured
CI: Confidence interval; ITBF: Iliotibial band friction syndrome (knee tendinitis); RR: Risk ratio.
a Randomization
and allocation concealment were unclear. Only assessors were blinded. No intention-to-treat analysis was done, and
baseline imbalance was reported.
b Very few participants.
18
DISCUSSION
Summary of main results

Evidence was insufficient to demonstrate a consistent clinically
important benefit of deep transverse friction massage combined
with additional physical therapy modalities when compared with a
control in the treatment of extensor carpi radialis tendinitis (lateral
elbow tendinitis) (Stratford 1989) and of iliotibial band friction
syndrome (knee tendinitis) (Schwellnus 1992). The additional
benefit of deep transverse friction massage was not confirmed, as
combined interventions were used. The included studies did not
report on adverse events.
Overall completeness and applicability of

evidence
Only two clinical trials met the inclusion criteria of this review update. Moreover, these two trials were at high risk of bias. Therefore
evidence on the specific benefits of deep transverse friction massage is sparse. Further, results of the intervention looking at deep
transverse friction massage and physical therapy versus physical
therapy alone may not be applicable to general populations, as participants in the trial were experienced runners. The Philadelphia
Panel recommends that evidence is insufficient (level 1, grade C for
pain) to include or exclude deep transverse friction massage alone
as a therapeutic intervention for treating tendinitis (Philadelphia
2001). Evaluation of deep transverse friction massage as a single
treatment in comparison with other techniques has not been undertaken, so its efficacy has not yet been demonstrated (Joseph
2012).
Confounding variables such as characteristics of the device, characteristics of the therapeutic application, characteristics of the population, characteristics of the disease, and methodological considerations might have contributed to the lack of effect (Morin 1996).
Characteristics of the technique described by Cyriax (Cyriax
1975a; Cyriax 1975b) such as years of experience of the therapist, characteristics of the application (pressure, rhythm and progression, and frequency), duration of treatment sessions and the
treatment schedule, characteristics of the population (age, gender),
characteristics of the disease (chronic vs acute conditions), and
weakness of methodological considerations (comparison groups,
sample size, study duration, nonvalidated outcome measures) in
both studies (Schwellnus 1992; Stratford 1989) may have contributed to inconclusive results on the effectiveness of deep transverse friction massage for tendinitis. FInally, larger and better powered studies are required to confirm findings because of the small
number of studies included in this update of the review and the
small sample sizes presented by the included trials.
For both studies, further research is very likely to have an important impact on our confidence in the estimate of effect and is
likely to change the estimate. In other words, we obtained very low
grades for the main outcomes, principally because of unclear allocation concealment and randomization, single-blind processes,
wide confidence intervals, and small sample size.
Potential biases in the review process

It is important to consider whether all relevant trials were identified (whether or not all pertinent data could be obtained) and if
the methods used (search strategy, selection of studies, data collection, analysis) could have introduced potential bias. When selecting studies for inclusion, we did not consider languages other
than French and English. This could have limited the number of
potentially eligible studies found.

For both included RCTs, the two major outcomes (pain and function) received a GRADE score of very low for the quality of the
body of evidence.
The two RCTs included in this systematic review highlight a common problem among trials of rehabilitation interventions: difficulty or inability to blind participants and personnel, which contributes to the high risk of bias. Although outcome assessors were
blinded in both studies, lack of blinding of participants may have
an impact on patient-reported outcomes such as pain, leading to
overestimation. Both presented baseline imbalance (mean age and
duration of symptoms were different across groups). One of the
studies (Schwellnus 1992) demonstrated high risk of incomplete
outcome data, as investigators reported a 15% dropout rate, and
no intention-to-treat analysis was performed. The small sample
sizes and wide confidence intervals also contribute to the very low
quality rating of the body of evidence for reported outcomes.
Agreements and disagreements with other

studies or reviews
This systematic review of deep transverse friction massage for tendinitis was initially conducted as part of a guideline development
project entitled the Philadelphia Panel Guidelines on Rehabilitation Interventions for knee pain (Philadelphia 2001). Other guidelines such as ACR 1996, ACR 2000, Godlee 2000, and Manal
1996 did not evaluate any type of massage as a treatment intervention for knee conditions. To our knowledge, no guidelines on
massage for extensor carpi radialis tendinitis are available. The
American Physical Therapy Association (APTA) guidelines (APTA
2001) recommend massage for musculoskeletal conditions, although the APTA guidelines do not differentiate between types
of massage. However, these guidelines are not based on evidence
from comparative controlled trials.
19
No meta-analysis or literature reviews have reported the efficacy of

this type of massage (Chapman 1991; Furlan 2008; Green 1998;
van der Heijden 1997). Evaluation of deep transverse friction massage as a single treatment in comparison with other techniques has
not been undertaken, so its efficacy has not been yet demonstrated.
Comparison of different studies is difficult to perform because of
heterogeneity in terms of (1) types of tendinitis, (2) confounding approaches, and (3) varied outcome measures selected (Joseph
2012).
In the included RCT (Stratford 1989), deep transverse friction
massage was performed to reduce symptoms of tendinitis. The inflammation and pain observed in tendinitis are frequently due to
three main factors: (1) biomechanical factors), (2) anthropometric
factors, and (3) training factors. Pain is an indirect symptom. Based
on identified factors, pain could be controlled more effectively
through other physical therapy interventions such as strengthening and postural exercises, or changes in functional and sporting activities that correct biomechanical deficiencies (Thaunton
1987), restore motion (Hart 1994), increase strength, endurance,
and function (Hart 1994; Thaunton 1987), and permit gradual return to training (Thaunton 1987). One included study (Stratford
1989) found that phonophoresis is not superior to ultrasound and
placebo ointment, where similar results were obtained. Antich et
al (Antich 1986) mentioned that phonophoresis is as good as ice
and ultrasound. Halle and colleagues (Halle 1986) confirmed that
ultrasound, transcutaneous electrical nerve stimulation (TENS),
and cortisone injections represent comparative therapies in terms
of efficacy. The second included RCT (Schwellnus 1992) suggested that the addition of deep transverse friction massage might
not appear to aggravate clinical outcomes in iliotibial band friction syndrome (knee tendinitis). Therefore, the included study rejected the findings of the first study author, who examined the biologic rationale of this therapy (Griffin 1963) and reported that the
success rate was significantly greater in the phonophoresis group
compared with the group given ultrasound therapy. Similar success rates for the phonophoresis groups were shown in the work of
Kleinkort et al (Griffin 62% vs Kleinkort 66%) (Stratford 1989).
However, it is important to note that those study authors did not
consider the presence of potential confounding variables, and neither of them reported their diagnostic criteria and the power of
their statistical test (Stratford 1989).
Additional well-designed RCTs are needed before this specific type
of massage can be excluded for treatment of this condition. Clinicians and patients need to be presented with evidence that will
enable them to make informed decisions regarding treatment options.
AUTHORS CONCLUSIONS
Implications for practice
Evidence was insufficient to reveal the clinically important benefit
of deep transverse friction massage (DTFM) in treating tendinitis. Unfortunately, almost no persuasive scientific research about
DTFM has been reported. The therapeutic approach consists
mainly of reasonable speculation about biology, as the effects of
DTFM are based on sound physiologic and pathologic theories.
In other words, many patients respond well to any kind of stimulation techniques that may alter a chronically painful condition.
Even though the technique seems to work well in practice, the
absence of evidence remains a matter of primary concern, as the
use of DTFM must absolutely be supported by clinical evidence.
Therefore, it is clear that better designed studies are needed before
conclusions can be drawn about the efficacy or lack of efficacy of
deep transverse friction massage for treating symptomatic tendinitis (Joseph 2012).
Implications for research

To justify the use of deep transverse friction massage in the treatment of tendinitis, large, high-quality randomized trials using validated outcome measures and high-quality reporting methods are
needed.
ACKNOWLEDGEMENTS
The review authors thank Lisa Levesque, Shaman Gibeault, Judith
Robitaille, Michel Boudreau, Michael Saginur, and Sarah Clment
for help with data extraction and literature retrieval, as well as the
editorial team of the Cochrane Musculoskeletal Review Group for
valuable comments on early drafts.
20
REFERENCES
References to studies included in this review

Schwellnus 1992 {published data only}
Schwellnus MP, Mackintosh L, Mee J. Deep transverse
frictions in the treatment of iliotibial band friction syndrome
in athletes: a clinical trial. Physiotherapy 1992;78(8):5648.
Stratford 1989 {published data only}
Stratford PW, Levy DR, Gauldie S, Miseferi D, Levy K.
The evaluation of phonophoresis and friction massage
as treatments for extensor carpi radialis tendinitis: a
randomized controlled trial. Physiotherapy Canada 1989;41
(2):939.
References to studies excluded from this review

Balke 1989 {published data only}
Balke B, Anthony J, Wyatt F. The effects of massage
treatment on exercise fatigue. Clinical Sports Medicine 1989;
1:18996.
Baltaci 2001 {published data only}
Baltaci G, Ergun N, Tunay VB. Effectiveness of Cyriax
manipulative therapy and elbow band in the treatment of
lateral epicondylitis. European Journal of Sports Traumatology
and Related Research 2001;23(3):1138.
Bisset 2007 {published data only}
Bisset L, Smidt N, Van der Windt DA, Bouter LM, Jull G,
Brooks P, et al. Conservative treatments for tennis elbow-do
subgroups of patients respond differently?. Rheumatology
2007;1:16015.
Chiarello 1997 {published data only}
Chiarello CM, Gundersen L, OHalloran T. The effect of
continuous passive motion duration and increment on range
of motion in total knee arthroplasty patients. Journal of
Orthopaedic & Sports Physical Therapy 1997;25(2):11927.
Crosman 1984 {published data only}
Crosman LJ, Chateauvert SR, Weisberg J. The effects of
massage to the hamstring muscle group on range of motion.
Journal of Orthopaedic and Sports Physical Therapy 1984;6:
16872.
Feehan 1989 {published data only}
Feehan RC. The efficacy of using transverse friction massage
on improving active and passive range of motion in the
client with chronic knee dysfunction. The Union Institute
1989;51(3):64.
Fernandez 2006 {published data only}
Fernndez-de-las-Penas C, Alonso-Blanco C, FernandezCarnero J, Miangolarr-Page J. The immediate effect of
ischemic compression technique and transverse friction
massage on tenderness of active and latent myofascial trigger
points: a pilot study. Journal of Bodywork and Movement
Therapies 2006;10:80711.
Fernandez 2008 {published data only}
Fernndez-Prez AM, Peralta-Ramrez MI, Pilat A,
Villaverde C. Effects of myofascial induction techniques
on physiologic and psychologic parameters: a randomized

controlled trial. The Journal of Alternative and
Complementary Medicine 2008;14(7):80711.
Iwatsuki 2001 {published data only}
Iwatsuki H, Ikuta Y, Shinoda K. Deep friction massage on
the masticatory muscles in stroke patients increases biting
force. Journal of Physical Therapy Science 2001;13(1):1720.
Joseph 2012 {published data only}
Joseph MF, Taft K, Moskwa M. Deep friction massage to
treat tendinopathy: a systematic review of a classic treatment
in the face of a new paradigm of understanding. Journal of
Sports Rehabilitation 2012;21(4):34353.
Kohia 2008 {published data only}
Kohia M, Brackle J, Byrd K, Jennings A, Murray W,
Wilfong E. Effectiveness of physical therapy treatment on
lateral epicondylitis. Journal of Sport Rehabilitation 2008;17
(2):11936.
Malier 1986 {published data only}
Malier M, Troisier O. Deep transverse massage for the
treatment of tennis elbow [La place du massage transversal
profond dans le traitement des tendinites picondyliennes:
propos de 131 cas]. Annales de Radaptation et de Mdecine
Physique 1986;29:7583.
Mayer 2007 {published data only}
Mayer F, Hirschmuller A, Muller S, et al. Effect of
short-term treatment strategies over 4 weeks in Achilles
tendinopathy. British Journal of Sport Medecine 2007;41(7):
e6.
Nagrale 2009 {published data only}
Nagrale AV, Herd CR, Ganvir S, Ramteke G. Cyriax
physical therapy versus phonophoresis with supervised
exercise in subjects with lateral epicondylalgia: a randomized
clinical trial. Journal of Manual and Manipulative Therapy
2009;17(3):1718.
Pellecchia 1994 {published data only}
Pellecchia GL, Hamel H, Behnke P. Treatment of
infrapatellar tendinitis: a combination of modalities and
transverse friction massage versus iontophoresis. Journal of
Sport Rehabilitation 1994;3(2):13545.
Smidt 2002 {published data only}
Smidt N, Van der Windt DA, Assendelft WJJ, Devill
WL, Korthals-de Bos IB, Bouter LM. Corticosteroid
injections, physical therapy, or a wait-and-see policy for
lateral epicondylitis: a randomised controlled trial. The
Lancet 2002;359:65762.
Stasinopoulos 2004a {published data only}
Stasinopoulos D, Johnson MI. Curiax physical therapy for
tennis elbow/lateral epicondylitis. British Journal of Sport
Medecine 2004;38(6):6757.
Stasinopoulos 2004b {published data only}
Stasinopoulos D, Stasinopoulos I. Comparison of effective
exercise programme, pulsed ultrasound and transverse
21
friction in the treatment of chronic patellar tendinopathy.

Clinical Rehabilitation 2004;18(4):34752.
Stasinopoulos 2006 {published data only}
Stasinopoulos D, Stasinopoulos I. Comparison of effects
of Cyriax physical therapy, a supervised exercise program
and polarized polychromatic non-coherent light (Bioptron
light) for the treatment of lateral epicondylitis. Clinical
Rehabilitation 2006;20(1):1223.
Struijs 2003 {published data only}
Struijs P, Damen PJ, Bakker E, Blankevoort L, Assendelft
WJ, van Dijk CN. Manipulation of the wrist for
management of lateral epicondylitis: a randomized pilot
study. Physical Therapy 2003;83(7):60816.
Struijs 2006 {published data only}
Struijs P, Korthals-de Bos I, van Tudler M, van Dijk CN,
Bouter LM, Assendelft WJ. Cost effectiveness of brace,
physical therapy, or both for treatment of tennis elbow.
British Journal of Sports Medecine 2006;40:63743.
Allander 1974
Allander E. Prevalence, incidence and remission rates
of some common rheumatic diseases or syndromes.
Scandinavian Journal of Rheumatology 1974;3:14553.
Antich 1986
Antich TJ, Randall CC, Westbrook RA, et al. Physical
therapy treatment of knee mechanism disorders: comparison
of four treatment modalities. Journal of Orthopaedic and
Sports Physical Therapy 1986;8:2559.
APTA 2001
American Physical Therapy Association. Guide to Physical
Therapist Practice: Part One: A Description of Patient/Client
Management. Alexandria, Va: American Physical Therapy
Association, 2001.
Chapman 1991
Chapman CE. Can the use of physical modalities for pain
control be rationalized by the research evidence?. Canadian
Journal of Physiology and Pharmacology 1991;69:70412.
Thomee 1997 {published data only}

Thomee RA. A comprehensive treatment approach for
patellofemoral pain syndrome in young women. Physical
Therapy 1997;77(12):1690703.
Cyriax 1975a
Cyriax J. Diagnosis of soft tissue lesions. In: Cyriax J editor
(s). Textbook of Orthopaedic Medicine. 9th Edition. Vol. 1,
Baltimore: Williams and Wilkins, 1975.
Verhaar 1995 {published data only}

Verhaar JAN, Walenkamp GH, van Mameren H, Kester
AD, van der Linden AJ. Local corticosteroid injection
versus Cyriax-type physical therapy for tennis elbow. Journal
of Bone and Joint Surgery [British Volume] 1995;77:1283.
Cyriax 1975b
Cyriax J. Treatment by manipulation, massage and injection.
In: Cyriax J editor(s). Textbook of Orthopaedic Medicine. 9th
Edition. Vol. 2, Baltimore: Williams and Wilkins, 1975.
Viswas 2012 {published data only}

Viswas R, Ramachandran R, Anantkumar P. Comparison
of effectiveness of supervised exercise program and Cyriax
physiotherapy in patients with tennis elbow (lateral
epicondylitis): a randomized clinical trial. Scientific World
Journal 2012;doi:10.1100/2012/939645.
Zhang 1987 {published data only}
Zhang ZB, Carter RM, Minikin BR, Telford RD. The
influence of repeated massage on leg strength. Australian
Institute of Sport Canberra 1987;5.
Zheng 2012 {published data only}
Zheng Z, Wang J, Gao Q, Hou J, Ma L, Jiang C, Chen
G. Therapeutic evaluation of lumbar tender point deep
massage for chronic non-specific low back pain. Journal of
Traditional Chinese Medicine 2014;32(4):5347.
Additional references
Furlan 2008
Furlan AD, Imamura M, Dryden T, Irvin E. Massage for low
back pain. Cochrane Database of Systematic Reviews 2001,
Issue 4. [DOI: 10.1002/14651858.CD001929.pub2]
Godlee 2000
Godlee Fiona (editor). Clinical Evidence: A Compendium
of the Best Available Evidence for Effective Health Care. First
Edition. Vol. 4, London, England: BMJ Publishing Group,
2000.
Green 1998
Green S, Buchbinder R, Glazier R, Forbes A. Systematic
review of randomised controlled trials of interventions for
painful shoulder: selection criteria, outcomes assessment,
and efficacy. BMJ 1998;316:35460.
Grewal 2009
Grewal R, MacDermid JC, Shah P, King GJ. Functional
outcome of arthroscopic extensor carpi radialis brevis
tendon release in chronic lateral epicondylitis. Journal of
Hand Surgery [American Volume] 2009;34(5):84957.
ACR 1996
American College of Rheumatology Ad Hoc Committee on
Clinical Guidelines. Guidelines for the management of
rheumatoid arthritis. Arthritis and Rheumatism 1996;39:
71322.
Griffin 1963
Griffin JE, Touchstone JC. Ultrasonic movement of cortisol
into pigs tissue. I.Movement into skeletal muscle. American
Journal of Physical Medicine 1963;42:7785.
ACR 2000
ACR: American College of Rheumatology Subcommittee
on Osteoarthritis Guidelines. Recommendations for the
medical management of osteoarthritis of the hip and knee.
Arthritis and Rheumatism 2000;43:190515.
Guyatt 2008
Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y,
Alonso-Coello P, et al. GRADE: an emerging consensus on
rating quality of evidence and strength of recommendations.
BMJ 2008;336:9246.
22
Halle 1986
Halle JS, Franklin RJ, Karalfa L. Comparison of four
treatment approaches for lateral epicondylitis of the elbow.
Journal of Orthopedic Sports Physical Therapy 1986;8:629.
Hart 1994
Hart LE. Exercises and soft tissue injury. Ballieres Clinical
Rheumatology 1994;8(1):13748.
Higgins 2011
Higgins JPT, Altman DG, Sterne JAC (editors). Chapter
8: Assessing risk of bias in included studies. In: Higgins
JPT, Green S (editors). Cochrane Handbook for Systematic
Reviews of Interventions [Version 5.1.0] [updated March
2011]. The Cochrane Collaboration, 2011. www.cochranehandbook.org.
Johnson 2007
Johnson GW, Cadwallader K, Scheffel SB, Epperly TD.
Treatment of lateral epicondylitis. American Family
Physician 2007;76(6):8438.
Jordaan 1994
Jordaan G, Schwellnus MP. The incidence of overuse
injuries in military recruits during basic military training.
Military Medicine 1994;159(6):1994.
Kirk 2000
Kirk KL, Kuklo T, Klemme W. Iliotibial band friction
syndrome: a review. Orthopedics 2000;23(11):120917.
Lavine 2010
Lavine R. Iliotibial band friction syndrome. Current Reviews
in Musculoskeletal Medicine 2010;3:1822.
Manal 1996
Manal RJ, Snyder-Mackler L, Manal RJ, Snyder-Mackler
L. Practice guidelines for anterior cruciate ligament
rehabilitation: criterion-based rehabilitation progression.
Operative Techniques in Orthopaedics 1996;6:1906.
Martens 1989
Martens M, Libbrecht P, Burssens A. Surgical treatment of
the iliotibial band friction syndrome. American Journal of
Sports Medicine 1989;17(5):6514.
Messier 1988
Messier SP, Pittala KA. Etiologic factors associated with
selected running injuries. Medicine & Science in Sports and
Exercise 1988;20(5):5015.
Messier 1995
Messier SP, Edwards DG, Martin DF, Lowery RB, Cannon
DW, James MK, et al. Etiology of iliotibial friction
syndrome in distance runners. Medicine and Sciences in
Sports and Exercises 1995;27(7):95160.
Morin 1996
Morin M, Brosseau L, Quirion-DeGrardi C. A theoretical
framework on low level laser therapy (classes I, II and III)
application for the treatment of OA and RA. Proceedings of
the Canadian Physiotherapy Association. 1996:1.
Philadelphia 2001
The Philadelphia Panel. The Philadelphia Panel
Evidence-Based Clinical Practice Guidelines on Selected
Rehabilitation Interventions for Knee Pain. Physical Therapy

2001;81(10):1300.
Pinshaw 1984
Pinshaw R, Atlas V, Noakes TD. The nature and response
to therapy of 196 consecutive injuries seen at a runners
clinic. South African Medical Journal 1984;65(8):2918.
Reinold 2002
Reinold MM, Wilk KE, Reed J, Crenshaw K, Andrews JR.
Interval sport programs: guidelines for baseball, tennis, and
golf. Journal of Orthopaedic and Sports Physical Therapy
2002;32(6):2938.
Roquelaure 2006
Roquelaure Y, Ha C, Leclerc A, Touranchet A, Sauteron
M, Melchior M, et al. Epidemiologic surveillance of
upper-extremity musculoskeletal disorders in the working
population. Arthritis and Rheumatism 2006;55(5):76578.
Schwellnus 1991
Schwellnus MP, Theunissen L, Noakes TD, Reinach SG.
Anti-inflammatory and combined anti-inflammatory/
analgesic medication in the early management of iliotibial
band friction syndrome: a clinical trial. South Africa Medical
Journal 1991;79(10):6026.
Schnemann 2011a
Schnemann HJ, Oxman AD, Vist GE, Higgins JPT,
Deeks JJ, Glasziou P, Guyatt GH. Chapter 12: Interpreting
results and drawing conclusions. In: Higgins JPT, Green
S (editors), Cochrane Handbook for Systematic Reviews
of Interventions [Version 5.1.0] [updated March 2011].
The Cochrane Collaboration, 2011. www.cochranehandbook.org.
Schnemann 2011b
Schnemann HJ, Oxman AD, Higgins JPT, Vist GE,
Glasziou P, Guyatt GH. Chapter 11: Presenting results
and Summary of findings tables. In: Higgins JPT, Green
S (editors), Cochrane Handbook for Systematic Reviews
of Interventions [Version 5.1.0] [updated March 2011].
The Cochrane Collaboration, 2011. www.cochranehandbook.org.
Shiri 2006
Shiri R, Viikari-Juntura E, Varonen H, Helio M. Prevalence
and determinants of lateral and medial epicondylitis: a
population study. American Journal of Epidemiology 2006;
164(11):1065-74.
Strauss 2011
Strauss EJ, Kim S, Calcei JG, Park D. Iliotibial band
syndrome: evaluation and management. Journal of the
American Academy of Orthopaedic Surgeons 2011;19(12):
72836.
Struijs 2002
Struijs PAA, Smidt N, Arola H, van Dijk CN, Buchbinder
R, Assendelt WJJ. Orthotics devices for the treatment of
tennis elbow. Cochrane Database of Systematic Reviews 2002,
Issue 1. [DOI: 10.1002/14651858.CD001821]
23
Thaunton 1987
Thaunton JE, Clement DB, Smart GW, McNicol KL. Nonsurgical management of overuse knee injuries in runners.
Canadian Journal of Sport Sciences 1987;12(1):1987.
van der Heijden 1997
van der Heijden GJ, van der Windt DA, de Winter AF,
van der Heijden GJ, van der Windt DA, de Winter
AF. Physiotherapy for patients with soft tissue shoulder
disorders: a systematic review of randomised clinical trials.
BMJ 1997;315(7099):2530.
Walker 1984
Walker JM. Deep transverse frictions in ligament healing.
Journal of Orthopaedic and Sports Physical Therapy 1984;6

(2):8994.
Walker-Bone 2004
Walker-Bone K, Palmer KT, Reading I, Coggon D, Cooper
C. Prevalence and impact of musculoskeletal disorders of
the upper limb in the general population. Arthritis and
Rheumatism 2004;51(4):64251.
Wilk 2004
Wilk KE, Reinold MM, Andrews JR. Rehabilitation of the
throwers elbow. Clinics in Sports Medicine 2004;23(4):765801, xii.
Indicates the major publication for the study
24
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]

Schwellnus 1992
Methods
Study design: randomized, assessor-blinded trial

Sample size at entry: 20
G1: deep transverse friction massage (DTFM) group 10
G2: control group 10
Treatment duration: 10 days
Follow-up: 14 days
Participants
Setting: visitors to a sports injury clinic with unilateral chronic (> 4 weeks) iliotibial band
syndrome causing pain severe enough to restrict running distance or speed (grade 3), or
to prevent it altogether (grade 4)
Inclusion criteria:
Age in years (mean ( SE)): DTFM group 25 ( 6), control group 29 ( 5), P
value 0.20 student t-test
Weeks injured (mean ( SE)): DTFM group 23 ( 17), control group 74 ( 95)
Years of running (mean ( SE)): DTFM group 7.7 ( 5.5), control group 5.4 ( 6.
2)
Kilometers run per week (mean ( SE)): DTFM group 45 ( 15), control group
64 ( 30)
Grade of injury (mean ( SE)): DTFM group 3.4 ( 0.5), control group 3.4 ( 0.
5)
Exclusion criteria: < 18 years old, history of previous knee surgery, concomitant medical
therapy
Interventions
Deep transverse friction massage: applied on days 3, 5, 7 and 10

Treated anatomical area (most tender area) with constant pressure, such that discomfort
was experienced, but not severe pain; 2 minutes of light friction, then 8 minutes of harder
friction
Technique: Pressure was applied with the index finger and reinforced with the ring finger,
with the thumb acting as a pivot; brisk motion was initiated from the therapists shoulder,
with the wrist flexible and the hand stiff
Concurrent treatment:
Rest (apart from 3 * 30 min treadmill exercise tests on days 3, 7 and 14)
Ice 20 minutes twice daily
Stretch of iliotibial band, daily
Ultrasound: 1 MHz, 0.5 W/cm/cm (continuous), 5 minutes on days 3, 4, 5 and
6, then 7 minutes on days 7 and 10
G1 received deep transverse friction massage and concurrent treatment while
control group received concurrent treatment only
Outcomes
Mean pain daily recall, total pain while running, % max pain experienced while running
(VAS 0 to 10; 0 = no pain) for 3 treatment periods (days 0 to 2, 3 to 6, 7 to 14)
25
Schwellnus 1992
(Continued)
Notes
One participant was excluded for refusal to comply with the treatment group. Two
participants were not accounted for in the control group (reasons not provided)
Per-protocol analysis
Source of funding: This research project was generously supported financially by Johnson and Johnson Pty Ltd
Declarations of interest of primary researchers not provided
Risk of bias
Bias
Authors judgement
Support for judgement
Random sequence generation (selection Unclear risk

bias)
Quote: patients were divided into two

treatment groups on a random basis
Comments: method not described
Allocation concealment (selection bias)
Comments: Investigators state that they divided the 17 athletes into 2 groups without
explaining the method of allocation
Unclear risk
Blinding of participants and personnel High risk

(performance bias)
All outcomes
Comment: Participants and personnel were

not blinded
Blinding of outcome assessment (detection High risk

bias)
All outcomes
Quote: the efficacy of the treatment was

assessed by a different therapist who was
blind to the treatment group of the patient
Comments: Insufficient information available to judge whether not blinding the
participants had an impact on patient-reported outcomes. More likely at high risk
for self-reported subjective outcomes (eg,
pain)
Incomplete outcome data (attrition bias)

All outcomes
High risk
Comments: This study had a 15% dropout

rate (even across treatment groups). Duration of follow-up was mentioned, but investigators did not perform an intentionto-treat analysis
Selective reporting (reporting bias)
Low risk
Comments: The trial presented planned

outcomes whether or not the result was significant. No selective reporting was found
to be present
Other bias
High risk
Comment: Baseline imbalance was reported; mean age and duration of symptoms were different across groups
26
Stratford 1989
Methods
Study design: randomized controlled trial, parallel group, combination of interventions

Sample size at entry: 40
Group 1: ultrasound and placebo ointment without frictions-9
Group 2: ultrasound and placebo ointment with frictions-11
Group 3: phonophoresis without frictions-10
Group 4: phonophoresis with frictions-10
# male/# female: 20/20
Group 1: 2/7
Group 2: 5/6
Group 3: 5/5
Group 4: 8/2
Treatment duration: 9 sessions within 5 weeks
Follow-up: 5 weeks
Participants
Setting: patients attending a community sports injuries clinic

Inclusion criteria: those who complained of discomfort at or about the lateral epicondyle;
pain at the lateral aspect of the elbow during resisted wrist extension; radial deviation
during complete elbow extension; tenderness in palpation over, or at, one of the following
areas: (1) origin of extensor carpi radialis longus tendon, (2) origin of extensor carpi
radialis brevis tendon, (3) extensor carpi radialis brevis at tendon body, and (4) extensor
carpi radialis brevis tendon with tenderness extending from origin to the tendon body
Age (mean in years ( SD))
Group 1: 43.8 ( 9.8)
Group 2: 44.6 ( 9.8)
Group 3: 40.1 ( 8.3)
Group 4: 44.7 ( 8.7)
Disease duration (months: mean ( SD))
Group 1: 4.3 ( 3.2)
Group 2: 2.1 ( 1.2)
Group 3: 5.2 ( 7.2)
Group 4: 5.4 ( 4.1)
Exclusion criteria: combined lesions; bilateral elbow problems at initial assessment; history of prior surgery; history of an injection to the elbow within the past 6 months
Interventions
Deep transverse friction massage

Treated anatomical area (elbow) with friction massage applied perpendicular to the
structure of interest for 10 minutes, 3 times per week, 9 treatment sessions within 5
weeks
Frictions for groups 2 and 4: position of participant: for lesion at origin of the extensor
carpi radialis longus or brevis tendon, elbow flexed at 90 degrees with forearm fully
supinated; if lesion at or included tendon body or extensor carpi radialis brevis tendon,
elbow flexed at 45 degrees with forearm pronated
Concurrent treatment for group 1 and group 2: ultrasound and placebo ointment
Ultrasound: Dosage varied from 1.3 w/cm2 continuous output to 0.5 w/cm2 pulsed (1:
4). Application technique: sound head moved in slow concentric circles, while sound
head contact with the participant was maintained at the same time. Duration: 6 minutes
Concurrent treatment for groups 3 and 4: phonophoresis
Phonophoresis: 10% hydrocortisone ointment used with ultrasound treatment
27
Stratford 1989
(Continued)
Outcomes
Pain
Pain-free function (8 pain-free item, 8 = better)
Pain VAS (0 to 100 mm; 0 = worst).
Grip strength
Ration index of pain-free grip strength (grip strength: kilograms of force). Ratio is
pain-free grip divided by maximum grip of uninvolved limb
Function
Function VAS (0 to 100 mm; 0 = worst)
Functional status (success or failure to perform pain-free strengthening program for the
wrist extensor muscles, with the elbow extended, without subsequent regression within
2 weeks of follow-up)
Notes
We extracted the outcome of pain on VAS (0 to 100), not pain-free function, for the
review
Source of funding: This work was supported by the Physiotherapy Foundation of
Canada
Declarations of interest of primary researchers not provided
Risk of bias
Bias
Authors judgement
Support for judgement
Random sequence generation (selection Low risk

bias)
Quote: The study design was a two by

two factorial design. The subjects were randomly assigned using a balanced blocked
randomization table
Allocation concealment (selection bias)
Comments: Method of allocation concealment was not described
Unclear risk
Blinding of participants and personnel High risk

(performance bias)
All outcomes
Comment: Participants and personnel were

not blinded to treatment
Blinding of outcome assessment (detection Unclear risk

bias)
All outcomes
Quote: Decision was made by the assessor

who was blind to both interventions
Comments: Insufficient information was
available on which to base judgement if not
blinding participants had an impact on patient-reported outcomes
Incomplete outcome data (attrition bias)

All outcomes
Low risk
Comments: No description of withdrawals

and dropouts was mentioned. However,
all randomly assigned participants were accounted for in the results
Selective reporting (reporting bias)
Low risk
Comments: The trial presented all planned

outcomes whether or not the result was significant. No selective reporting was found
28
Stratford 1989
(Continued)
to be present
Other bias
High risk
Comments: Baseline imbalance was reported; the duration of symptoms was different across groups
DTFM: Deep transverse friction massage.

SD: Standard deviation.
SE: Standard error.
VAS: Visual analog scale.
Characteristics of excluded studies [ordered by study ID]
Study
Reason for exclusion
Balke 1989
Not tendinitis
Baltaci 2001
Combination of interventions; co-interventions not applied to the control group
Bisset 2007
Combination of Interventions; co-interventions not applied to the control group
Chiarello 1997
Healthy participants
Crosman 1984
Healthy participants
Feehan 1989
Not tendinitis
Fernandez 2006
Fernandez 2008
Iwatsuki 2001
Study was not a randomized controlled trial
Joseph 2012
Kohia 2008
No control group
Malier 1986
No control group
Mayer 2007
No control group
Nagrale 2009
29
(Continued)
Pellecchia 1994
Smidt 2002
Stasinopoulos 2004a
Stasinopoulos 2004b
No control group
Stasinopoulos 2006
Combination of interventions: Cyriax and Mills manipulation; co-intervention not applied to the control
group
Struijs 2003
Pilot study
Struijs 2006
Thomee 1997
Not tendinitis
Verhaar 1995
No control group
Viswas 2012
Combination of interventions: Cyriax and Mills manipulation; co-intervention not applied to the control
group
Zhang 1987
Not tendinitis
Zheng 2012
30
DATA AND ANALYSES
Comparison 1. Massage + therapeutic ultrasound (US) and placebo ointment vs US + placebo ointment only
Outcome or subgroup title

1 Pain (VAS 0-100, 0 = worst)
(change from baseline)
2 Grip strength (ratio index,
higher is better)
3 Function (VAS 0-100, 0 = worst)
4 Function (pain-free function;
average number of pain-free
items; higher is better)
5 Functional status (number
of successes to perform
strengthening program)
No. of
studies
No. of
participants
Statistical method
Effect size
Mean Difference (IV, Fixed, 95% CI)
Totals not selected
Totals not selected
1
1

Totals not selected

Totals not selected
Risk Ratio (M-H, Fixed, 95% CI)
Totals not selected
Comparison 2. Massage + phonophoresis vs phonophoresis only

1 Pain (VAS 0-100, 0 = worst)
2 Grip strength (ratio index,
higher is better)
3 Function (VAS 0-100, 0 = worst)
4 Function (pain-free function;
average number of pain-free
items; higher is better)
5 Functional status (number
of successes to perform
strengthening program)
No. of
studies
No. of
participants
Statistical method
Effect size
1
1

Totals not selected

Totals not selected
1
1

Totals not selected

Totals not selected
Risk Ratio (M-H, Fixed, 95% CI)
Totals not selected
31
Comparison 3. Massage + physical therapy vs physical therapy only
No. of
studies

1 Pain
1.1 Daily pain
1.2 Pain while running
1.3 % of maximum pain while
running
No. of
participants
Statistical method
1
1
1
1
Effect size

Totals not selected

0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
0.0 [0.0, 0.0]
Analysis 1.1. Comparison 1 Massage + therapeutic ultrasound (US) and placebo ointment vs US + placebo
ointment only, Outcome 1 Pain (VAS 0-100, 0 = worst) (change from baseline).
Review:
Deep transverse friction massage for treating lateral elbow or lateral knee tendinitis
Comparison: 1 Massage + therapeutic ultrasound (US) and placebo ointment vs US + placebo ointment only
Outcome: 1 Pain (VAS 0-100, 0 = worst) (change from baseline)
Study or subgroup
Stratford 1989
Treatment
Mean
Difference
Control
Mean(SD)
Mean(SD)
11
-11.2 (30.2)
-4.6 (19.7)
Mean
Difference
IV,Fixed,95% CI
IV,Fixed,95% CI
-6.60 [ -28.60, 15.40 ]
-100
-50
Favours treatment
50
100
Favours control
32
ointment only, Outcome 2 Grip strength (ratio index, higher is better).
Review:
Outcome: 2 Grip strength (ratio index, higher is better)
Study or subgroup
Stratford 1989
Treatment
Mean
Difference
Control
Mean(SD)
Mean(SD)
11
0.7 (0.3)
0.6 (0.3)
Mean
Difference
IV,Fixed,95% CI
IV,Fixed,95% CI
0.10 [ -0.16, 0.36 ]
-1
-0.5
0.5
Favours control
Favours treatment
ointment only, Outcome 3 Function (VAS 0-100, 0 = worst).
Review:
Outcome: 3 Function (VAS 0-100, 0 = worst)
Study or subgroup
Stratford 1989
Treatment
Mean
Difference
Control
Mean(SD)
Mean(SD)
11
76.3 (21.9)
78.1 (16.5)
Mean
Difference
IV,Fixed,95% CI
IV,Fixed,95% CI
-1.80 [ -18.64, 15.04 ]
-100
-50
Favours control
50
100
Favours treatment
33
ointment only, Outcome 4 Function (pain-free function; average number of pain-free items; higher is better).
Review:
Outcome: 4 Function (pain-free function; average number of pain-free items; higher is better)
Study or subgroup
Stratford 1989
Treatment
Mean
Difference
Control
Mean(SD)
Mean(SD)
11
3.8 (2.7)
2.7 (2.1)
Mean
Difference
IV,Fixed,95% CI
IV,Fixed,95% CI
1.10 [ -1.00, 3.20 ]
-4
-2
Favours control
Favours treatment
ointment only, Outcome 5 Functional status (number of successes to perform strengthening program).
Review:
Outcome: 5 Functional status (number of successes to perform strengthening program)
Study or subgroup
Stratford 1989
Treatment
Control
n/N
n/N
4/11
1/9
Risk Ratio
Risk Ratio
M-H,Fixed,95% CI
M-H,Fixed,95% CI
3.27 [ 0.44, 24.34 ]
0.01
0.1
Favours control
10
100
Favours treatment
34
Analysis 2.1. Comparison 2 Massage + phonophoresis vs phonophoresis only, Outcome 1 Pain (VAS 0-100, 0
= worst).
Review:
Comparison: 2 Massage + phonophoresis vs phonophoresis only

Outcome: 1 Pain (VAS 0-100, 0 = worst)
Study or subgroup
Stratford 1989
Treatment
Mean
Difference
Control
Mean(SD)
Mean(SD)
10
-0.2 (18.85)
10
1 (24.25)
Mean
Difference
IV,Fixed,95% CI
IV,Fixed,95% CI
-1.20 [ -20.24, 17.84 ]
-100
-50
50
Favours treatment
100
Favours control
Analysis 2.2. Comparison 2 Massage + phonophoresis vs phonophoresis only, Outcome 2 Grip strength
(ratio index, higher is better).
Review:

Outcome: 2 Grip strength (ratio index, higher is better)
Study or subgroup
Stratford 1989
Treatment
Mean
Difference
Control
Mean(SD)
Mean(SD)
10
0.5 (0.3)
10
0.7 (0.3)
Mean
Difference
IV,Fixed,95% CI
IV,Fixed,95% CI
-0.20 [ -0.46, 0.06 ]
-0.5
-0.25
Favours control
0.25
0.5
Favours treatment
35
Analysis 2.3. Comparison 2 Massage + phonophoresis vs phonophoresis only, Outcome 3 Function (VAS 0100, 0 = worst).
Review:

Outcome: 3 Function (VAS 0-100, 0 = worst)
Study or subgroup
Stratford 1989
Treatment
Mean
Difference
Control
Mean(SD)
Mean(SD)
10
82.5 (16.3)
10
78.8 (23.7)
Mean
Difference
IV,Fixed,95% CI
IV,Fixed,95% CI
3.70 [ -14.13, 21.53 ]
-100
-50
50
Favours control
100
Favours treatment
Analysis 2.4. Comparison 2 Massage + phonophoresis vs phonophoresis only, Outcome 4 Function (painfree function; average number of pain-free items; higher is better).
Review:

Outcome: 4 Function (pain-free function; average number of pain-free items; higher is better)
Study or subgroup
Stratford 1989
Treatment
Mean
Difference
Control
Mean(SD)
Mean(SD)
10
3.7 (2.8)
10
3.6 (2.6)
Mean
Difference
IV,Fixed,95% CI
IV,Fixed,95% CI
0.10 [ -2.27, 2.47 ]
-4
-2
Favours control
Favours treatment
36
Analysis 2.5. Comparison 2 Massage + phonophoresis vs phonophoresis only, Outcome 5 Functional status
(number of successes to perform strengthening program).
Review:

Outcome: 5 Functional status (number of successes to perform strengthening program)
Study or subgroup
Treatment
Control
n/N
n/N
2/10
3/10
Stratford 1989
Risk Ratio
Risk Ratio
M-H,Fixed,95% CI
M-H,Fixed,95% CI
0.67 [ 0.14, 3.17 ]
0.01
0.1
Favours control
10
100
Favours treatment
Analysis 3.1. Comparison 3 Massage + physical therapy vs physical therapy only, Outcome 1 Pain.
Review:
Comparison: 3 Massage + physical therapy vs physical therapy only

Outcome: 1 Pain
Study or subgroup
Treatment
Mean
Difference
Control
Mean
Difference
Mean(SD)
Mean(SD)
IV,Fixed,95% CI
IV,Fixed,95% CI
0.6 (0.3)
1 (0.5)
-0.40 [ -0.80, 0.00 ]
20 (9)
23 (8)
-3.00 [ -11.08, 5.08 ]
8 (5)
8.1 (3)
-0.10 [ -3.97, 3.77 ]
1 Daily pain
Schwellnus 1992
2 Pain while running
Schwellnus 1992
3 % of maximum pain while running
Schwellnus 1992
-10
-5
Favours Treatment
10
Favours Control
37
ADDITIONAL TABLES
Table 1. Inclusion and exclusion critieria according to the PICOTS strategy
Inclusion
Exclusion
Participants/Population (P)
Outpatients or inpatients
Diagnosis: tendinitis pain
Chronic versus acute conditions
Normal weight (BMI < 25 kg/m2 )
Age groups 18 years
Medically stable
Mentally competent
Participants/Population (P)
Cancer (and other oncologic conditions)
Dermatologic conditions
Healthy normal
Mixed population
Multiple conditions (presenting other chronic problems
additional)
Neurologic conditions
Pediatric conditions
Psychiatric conditions
Pulmonary conditions
Scoliosis
Condition in which rapid weight loss or exercise is
contraindicated (angina, frailty, advanced osteoporosis)
Obese or overweight patient (BMI 25 kg/m2 )
Interventions (I)
Eligible interventions: deep transverse frictions techniques
only, in community or not, and with or without:
Concurrent programs (eg, stretching exercises,
modalities (ultrasound), phonophoresis)
Supervision
Eligible control groups: conventional therapy, untreated,
waiting list, active physical therapy treatments, educational
pamphlets
Interventions (I)
Surgery (ie, not the effects of surgery)
Medication (eg, phonophoresis with medications)
Thermal biofeedback
Comparisons (C)
Studies were included if they compared an intervention group (eg,
deep transverse frictions techniques combined with modalities,
exercises) with a comparison group (eg, placebo, no treatment,
active treatment such as modalities, exercises)
Comparisons (C)
Studies were excluded if they did not compare the intervention
group with a comparison group (eg, placebo, no treatment, active
treatment such as modalities, exercises)
Outcomes (O)
Functional status (self-care activities)
Medication intake (if reported)
Muscle strength
Pain intensity
Participant satisfaction
Quality of life
Compliance
Outcomes (O)
Biochemical measures
Participant compliance with medication
Psychosocial measures (depression, home and community
activities, leisure, social roles, sexual functions)
Serum markers (except ESR)
Period of time (P)

Period of time (P)
Studies were included if the intervention period lasted longer than Studies were excluded if the intervention period lasted less than 1
1 week or 1 treatment session, with or without a follow-up period week or 1 treatment session
38
Table 1. Inclusion and exclusion critieria according to the PICOTS strategy
Study designs (S)

Randomized controlled trial
Controlled clinical trial
*English and French articles only.
(Continued)
Study designs (S)

Case series/case report
Case control studies
Cohort studies
Data (graphic) without a mean and SD
Sample size of fewer than 5 participants per experimental
group
Studies with greater than 20% dropout rate
BMI: Body mass index.

ESR: Erythrocyte sedimentation rate.
PICOTS: Populations, interventions, comparators, outcomes, timing, and setting framework.
SD: Standard deviation.
APPENDICES
Appendix 1. MEDLINE search strategy
Ovid MEDLINE(R) In-Process & Other Nonindexed Citations, and Ovid MEDLINE(R) (1946 to July 1, 2014)
1 exp Tendinopathy/ (6229)
2 tendinopath$.tw. (1352)
3 tend?nit$.tw. (2331)
4 tend?nos$.tw. (576)
5 ECRT.tw. (11)
6 tennis elbow.tw. (650)
7 lateral$ epicondylitis.tw. (507)
8 ITBFS.tw. (18)
9 Iliotibial band friction syndrome.tw. (55)
10 exp Cumulative Trauma Disorders/ (10274)
11 (repetiti$ adj (motion disorder or strain)).tw. (358)
12 cumulative trauma disorder$.tw. (273)
13 or/1-12 (18667)
14 exp Musculoskeletal Manipulations/ (11415)
15 physical therapy modalities/ (26485)
16 massag$.tw. (6837)
17 Deep transverse friction.tw. (6)
18 dtfm.tw. (2)
19 Cyriax.tw. (31)
20 or/14-19 (41129)
21 13 and 20 (534)
39
Appendix 2. EMBASE search strategy

Database: EMBASE Classic + EMBASE (1947 to July 1, 2014)
1 exp massage/ (10894)
2 therapeutic touch.mp. (429)
3 reflexotherapy.mp. (416)
4 exp ROLFING/ (12)
5 exp SHIATSU/ (65)
6 exp reflexology/ (374)
7 myotherapy.mp. (34)
8 (polarity adj therapy).mp. (13)
9 (myofascial adj release).mp. (114)
10 (craniosacral adj therapy).mp. (87)
11 exp REIKI/ (184)
12 (trager adj psychophysical).mp. (6)
13 (hakomi adj method).mp. (2)
14 (jin adj shin).mp. (4)
15 (neuromuscular adj therapy).mp. (10)
16 (pfrimmer adj25 therapy).mp. (4)
17 (alexander adj technique).mp. (85)
18 exp Alexander Technique/ (35)
19 (feldenkrais adj method).mp. (57)
20 massag$.mp. (18532)
21 or/1-20 (19890)
22 exp tendon injury/ (15552)
23 tendinitis/ (6184)
24 tenosynovitis/ (4011)
25 tendinopathy/ (5701)
26 (tend#nitis or tend#nosis or tend#nopathy or tenovaginitis or paratend#nitis or peritend#nitis or tenosynovitis).tw. (7635)
27 or/22-26 (25837)
28 21 and 27 (172)
Appendix 3. CENTRAL search strategy

Wiley Cochrane Library - CENTRAL
Issue 7 of 12, July 2014
#1
MeSH descriptor: [Tendinopathy] explode all trees
#2
MeSH descriptor: [Tendon Injuries] explode all trees
#3
(Tendinitis or Tendinosis or Tendonitis):ti,ab
#4
lateral epicondylitis
#5
Iliotibial band friction syndrome
#6
repetitive motion
#7
repetitive strain
#8
#1 or #2 or #3 or #4 or #5 or #6 or #7
#9
Deep transverse friction
#10
massage
#11
MeSH descriptor: [Massage] explode all trees
#12
MeSH descriptor: [Musculoskeletal Manipulations] explode all trees
#13
dtfm
#14
Cyriax
#15
MeSH descriptor: [Physical Therapy Modalities] explode all trees
#16
#9 or #10 or #11 or #12 or #13 or #14 or #15
#17
#8 and #16
40
Appendix 4. CINAHL search strategy

EBSCO Host CINAHL
1982-July 2014
S15
S9 and S14
120
S14
S10 or S11 or S12 or S13
9452
S13
TX massage
9452
S12
(MH Deep Tissue Massage)
44
S11
(MH Sports Massage)
166
S10
(MH Massage)
5733
S9
S1 or S2 or S3 or S4 or S5 or S6 or S7 or S8
3869
S8
tendinitis
407
S7
repetitive strain
157
S6
(MH Cumulative Trauma Disorders)
1667
S5
ECRT
S4
lateral epicondylitis
240
S3
(MH Elbow Injuries)
579
S2
(MH Tennis Elbow)
645
S1
(MH Tendinopathy)
Appendix 5. PEDro search strategy

PEDro
Therapy: stretching, mobilization, manipulation and massage
Problem: pain
Body part: forearm elbow
41
Appendix 6. Clinicaltrials.gov search strategy

Clinicaltrials.gov
Conditions: tendinitis OR tenosynovitis OR tendinopathy
Interventions: massage
WHATS NEW
Last assessed as up-to-date: 1 July 2014.
Date
Event
Description
1 July 2014
New citation required but conclusions have not changed
New review authors
1 July 2014
New search has been performed
New search; no new studies

Added risk of bias tables, summary of findings table and
new excluded articles
CONTRIBUTIONS OF AUTHORS
LB and LML were responsible for writing the manuscript, extracting and analyzing data, and selecting trials for the initial review.
LML, PR and GDA performed data extraction and updated selections from the reference list, the analyses, and the interpretation of
results.
PR was responsible for the literature search update.
BS, PT, GW, VW and SP contributed methodological expertise and commented on early drafts.
DECLARATIONS OF INTEREST
All the authors have no conflict of interest to declare.
SOURCES OF SUPPORT
Internal sources
Institute for Population Health, University of Ottawa, Canada.
Ottawa Health Research Institute, Canada.
42
External sources
Holistic Health Research Foundation of Canada, Canada.
DIFFERENCES BETWEEN PROTOCOL AND REVIEW

We used the Risk of bias tool to assess the risk of bias in included studies and presented the results in Summary of findings tables.
INDEX TERMS
Medical Subject Headings (MeSH)
Combined Modality Therapy; Cryotherapy; Iliotibial Band Syndrome [ therapy]; Massage [ methods]; Ointments [administration &
dosage]; Phonophoresis; Randomized Controlled Trials as Topic; Rest; Tennis Elbow [ therapy]; Ultrasonic Therapy
MeSH check words

Humans
43

Deep Transverse Friction Massage For Treating Lateral Elbow or Lateral Knee Tendinitis

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Deep Transverse Friction Massage For Treating Lateral Elbow or Lateral Knee Tendinitis

Încărcat de

Drepturi de autor:

Formate disponibile

Cochrane Database of Systematic Reviews

Deep transverse friction massage for treating lateral elbow or

Deep transverse friction massage for treating lateral elbow or

PLAIN LANGUAGE SUMMARY

Illustrative comparative risks* (95% CI)

Quality of the evidence

Control (US + placebo Massage + US and

Mean change in pain in

Mean function (VAS 0100, 0 = worst) in the

Mean function (vas 0 to

Patient global assess- See comment

Adverse events not mea- See comment

Withdrawals due to ad- See comment

Description of the condition

2002; Thaunton 1987; Wilk 2004); (5) prevention of reinjury

How the intervention might work

Why it is important to do this review

Description of the intervention

Criteria for considering studies for this review

selected because they are defined as comparative controlled trials

Mean (or mean change in) pain.

In addition to these outcomes, one of the review authors (LB) has

Search methods for identification of studies

Figure 1. Study flow diagram.

Searching other resources

Data collection and analysis

Measures of treatment effect

Individual participants represented the unit of analysis in these

Data extraction and management

Dealing with missing data

The procedure was to contact the study authors if only an abstract

Assessment of risk of bias in included studies

Unit of analysis issues

Assessment of reporting biases

Subgroup analysis and investigation of heterogeneity

Grading of the evidence and summary of findings

Results of the search

Risk of bias in included studies

In the RCT for extensor carpi radialis tendinitis (lateral elbow

difference was not clinically or statistically significant. The quality

Deep transverse friction masssage + phonophoresis

massage and phonophoresis compared with 78.8 23.7 for

Deep transverse friction masssage and physical

while running measured as the total daily pain of all participants

Illustrative comparative risks* (95% CI)

Quality of the evidence

MD = -1.2 (-20.24 to 17.

MD = 3.70 (-14.13 to 21.

Control (phonophore- Massage

Mean change in pain in

Mean function in the in- tervention groups was

Patient global assess- See comment

Adverse events not mea- See comment

Withdrawals due to ad- See comment

Illustrative comparative risks* (95% CI)

Deep transverse massage

Mean daily pain

Quality of the evidence

Function not measured

Quality of life not mea- See comment

Patient global assess- See comment

Adverse events not mea- See comment

Withdrawals due to ad- See comment